{"paper_id":"41a97622-09c4-4b89-b47c-bd21b5741cd1","body_text":"Abstract\nObjective\nTo assess the diagnostic confidence of multiple readers in the magnetic resonance imaging (MRI) diagnosis of endometriosis.\nMethods\nSixty-five patients (mean age 33; range 19–45 years) who had undergone MRI were retrospectively evaluated. Five regions were analysed and the presence of endometriosis was scored on a five-point scale in order to assess the diagnostic confidence. Statistical analysis included receiver operating characteristic (ROC) curve analysis, the Cohen weighted test and sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), accuracy, positive likelihood ratio (LR+) and negative likelihood ratio (LR–).\nResults\nThe areas under the curve (AUC) in the detection of ovarian endometrioma were 0.942, 0.893 and 0.883 for readers 1, 2 and 3, respectively; in the uterosacral ligament (USL) AUCs were 0.907, 0.804 and 0.842; in the vaginal fornix (VF) 0.819, 0.733 and 0.69; in the anterior compartment 0.916, 0.833 and 0.873; and in the rectum/sigma/pouch of Douglas (RSD) 0.936, 0.856 and 0.834.\nConclusions\nDiagnostic confidence of the observers is different according to the region of the nodules of endometriosis and it can be challenging in the VF and for the less experience readers also in the AC and RSD. Moreover the degree of uncertain diagnosis for the less expert readers may reach up to one third of the examinations.\nKey points\n• Magnetic resonance imaging (MRI) is increasingly used to assess endometriosis\n• The diagnostic confidence of observers varies according to the location of endometriosis\n• The diagnosis is more difficult to establish by MRI in some anatomical locations\n• Specific training should be given concerning those locations that cause difficulty\nSimilar content being viewed by others\nReferences\nGiudice LC (2010) Clinical practice. Endometriosis. N Engl J Med 362:2389–2398\nBulun SE (2009) Endometriosis. N Engl J Med 360:268–279\nMacario S, Chassang M, Novellas S et al (2012) The value of pelvic MRI in the diagnosis of posterior cul-de-sac obliteration in cases of deep pelvic endometriosis. AJR Am J Roentgenol 199:1410–1415\nSiegelman ES, Oliver ER (2012) MR imaging of endometriosis: ten imaging pearls. Radiographics 32:1675–1691\nGuerriero S, Ajossa S, Garau N, Alcazar JL, Mais V, Melis GB (2010) Diagnosis of pelvic adhesions in patients with endometrioma: the role of transvaginal ultrasonography. Fertil Steril 94:742–746\nSaba L, Guerriero S, Sulcis R et al (2012) MRI and \"tenderness guided\" transvaginal ultrasonography in the diagnosis of recto-sigmoid endometriosis. J Magn Reson Imaging 35:352–360\nBazot M, Gasner A, Ballester M, Daraï E (2011) Value of thin-section oblique axial T2-weighted magnetic resonance images to assess uterosacral ligament endometriosis. Hum Reprod 26:346–353\nScardapane A, Bettocchi S, Lorusso F et al (2011) Diagnosis of colorectal endometriosis: contribution of contrast enhanced MR-colonography. Eur Radiol 21:1553–1563\nBusard MP, Mijatovic V, van Kuijk C, Hompes PG, van Waesberghe JH (2010) Appearance of abdominal wall endometriosis on MR imaging. Eur Radiol 20:1267–1276\nChassang M, Novellas S, Bloch-Marcotte C et al (2010) Utility of vaginal and rectal contrast medium in MRI for the detection of deep pelvic endometriosis. Eur Radiol 20:1003–1010\nSaba L, Guerriero S, Sulcis R, Ajossa S, Melis G, Mallarini G (2010) Agreement and reproducibility in identification of endometriosis using magnetic resonance imaging. Acta Radiol 51:573–580\nSaba L, Guerriero S, Sulis R et al (2011) Learning curve in the detection of ovarian and deep endometriosis by using magnetic resonance: comparison with surgical results. Eur J Radiol 79:237–244\nNg CS, Palmer CR (2007) Analysis of diagnostic confidence and diagnostic accuracy: a unified framework. Br J Radiol 80:152–160\nBrealey SD, Scally AJ, Hahn S, Godfrey C (2007) Evidence of reference standard related bias in studies of plain radiograph reading performance: a meta-regression. Br J Radiol 80:406–413\nRennie D (2003) Improving reports of studies of diagnostic tests: the STARD initiative. JAMA 289(1):89–90\nPark SH, Goo JM, Jo CH (2004) Receiver operating characteristic (ROC) curve: practical review for radiologists. Korean J Radiol 5:11–18\nZawin M, McCarthy S, Scoutt L (1989) Endometriosis: appearance and detection at MR imaging. Radiology 171:693–696\nZacharia TT, O’Neill MJ (2006) Prevalence and distribution of adnexal findings suggesting endometriosis in patients with MR diagnosis of adenomyosis. Br J Radiol 79:303–307\nBazot M, Darai E, Hourani R et al (2004) Deep pelvic endometriosis: MR imaging for diagnosis and prediction of extension of disease. Radiology 232:379–389\nRedwine DB (1987) The distribution of endometriosis in the pelvis by age groups and fertility. Fertil Steril 47:173–175\nKinkel K, Chapron C, Balleyguier C (1999) Magnetic resonance imaging characteristics of deep endometriosis. Hum Reprod 14:1080–1086\nGuerriero S, Alcázar JL, Ajossa S, Pilloni M, Melis GB (2009) Three-dimensional sonographic characteristics of deep endometriosis. J Ultrasound Med 28:1061–1066\nAcknowledgments\nThe authors would like to thanks Mrs Diana Caravana, Mr Franco Conti and Mr Paolo Contu for their precious help.\nAuthor information\nAuthors and Affiliations\nCorresponding author\nRights and permissions\nAbout this article\nCite this article\nSaba, L., Sulcis, R., Melis, G.B. et al. Diagnostic confidence analysis in the magnetic resonance imaging of ovarian and deep endometriosis: comparison with surgical results. Eur Radiol 24, 335–343 (2014). https://doi.org/10.1007/s00330-013-3013-9\nReceived:\nRevised:\nAccepted:\nPublished:\nIssue date:\nDOI: https://doi.org/10.1007/s00330-013-3013-9","source_license":"CC0","license_restricted":false}